Developing an advanced diagnostic model for hepatocellular carcinoma through multi-omics integration leveraging...

Area of Science:

• Oncology
• Genomics
• Molecular Biology

Background:

• Hepatocellular carcinoma (HCC) lacks effective diagnostic methods and etiological understanding.
• Programmed cell death-associated genes (CDRGs) play crucial roles in cancer development.
• Single-cell RNA sequencing (scRNA-seq) and RNA sequencing (RNA-seq) offer powerful tools for molecular profiling.

Purpose of the Study:

• To elucidate the role of CDRGs in HCC.
• To construct a diagnostic model for HCC using multi-omics data.
• To identify potential biomarkers and therapeutic targets for HCC.

Main Methods:

• Collected RNA-seq data from blood exosomes (exoRBase), cancer tissues (TCGA), and scRNA-seq data (GEO).
• Integrated differentially expressed genes (DEGs) from HCC cohorts with six categories of CDRGs.
• Employed machine learning and clinical indicators to develop a seven-gene diagnostic model.
• Utilized scRNA-seq and spatial transcriptome sequencing (stRNA-seq) to explore mechanisms and predict response to immune checkpoint blockade (ICB) therapy.
• Validated key gene expression via quantitative Polymerase Chain Reaction (qPCR) and immunohistochemistry.

Main Results:

• A seven-gene diagnostic model for HCC was established with high diagnostic efficacy (AUC up to 0.976).
• HCC cluster 3 showed higher stemness and microenvironment interactions, potentially representing HCC cell differentiation origins.
• Elevated expression of TRIB3 and NQO1 was observed in HCC cells, tissues, and exosomes.
• High TRIB3 and NQO1 expression may correlate with better response to ICB therapy.

Conclusions:

• The seven-gene model accurately screens for HCC.
• TRIB3 is a promising diagnostic biomarker and potential therapeutic target for HCC.
• Understanding CDRG roles can advance HCC diagnosis and treatment strategies.